OBJECTIVE: This project investigates whether supplementation with vitamin E or beta-carotene reduces incidence of lung, prostate, and other cancers, and tests several hypotheses related to cancer prevention, etiology (including gene-environment interactions), and early detection in a large, prospective cohort. BACKGROUND: Research indicates a role in cancer prevention for certain micronutrients, including vitamin E and beta-carotene, based on antioxidant, antiproliferative, and other properties. METHODS: The ATBC Study is a randomized, controlled intervention trial of daily supplementation with beta-carotene (20 mg) and/or vitamin E (50 mg dl-alpha-tocopheryl acetate) for 5-8 years. The 29,133 participants are 50-69 year old male cigarette smokers in southwestern Finland. Serum, toenails, and questionnaire data were collected at baseline, and serum, whole blood, and RBC's during followup. All cancers are identified annually through the national cancer registry, with central medical record and pathology reviews. PROGRESS: Active intervention concluded in 1993, and continued follow-up of the cohort examines post-trial incidence trends. Cohort and nested case-control investigations are underway, including many aimed at testing environmental-gene (ie, polymorphism) hypotheses. Intervention findings showed reduced incidence in the vitamin E group for prostate (32%) and colorectal (22%) cancers, and modest reductions in major cardiovascular events. Beta-carotene demonstrated no cancer preventive effects and resulted in increased lung cancer incidence (16%) and overall mortality (8%). Organ-specific intervention and cancer etiologic analyses are focused on a wide range of hypotheses that utilize the randomized interventions, baseline risk factor data, and prospectively collected biospecimens (serum, DNA, etc.). The research questions include, for example, effects of vitamin E, selenium, carotenoids, folate/homocysteine, vitamin D, alcohol, and other dietary components, steroid hormones, physical activity, and serum micronutrient concentrations on cancer. Also being investigated are mechanisms through which the two trial agents acted (and vitamin E in particular, given the prostate chemopreventive effect observed), and the cancer risk relation of several genetic polymorphisms and DNA markers including those for GSTs, CYPs, NQO1, folate and methyl-group metabolism, IGF's, androgen metabolism, vitamin D receptors, and DNA damage and repair.